Identifying high risk groups to prevent suicide - The ...nsrf.ie/wp-content/uploads/presentations/Prof Nav Kapur 11-09-14.pdfprevent suicide - The National Confidential Inquiry into

Identifying high risk groups to prevent suicide - The National

Confidential Inquiry into Suicide and Homicide in the UK

Professor Nav KapurThe Centre for Suicide Prevention University of Manchester

Cork, December 2008

National Confidential Inquiry into Suicide

Alison Roscoe Alyson Ashton Anna PearsonCathryn Rodway Damian da Cruz David While Harriet Bickley Huma Daud Isabelle HuntJames Burns Jenny Shaw Kelly HadfieldKirsten Windfur Louis Appleby Nav KapurNicola Swinson Pauline Turnbull Philip StonesPooja Saini Rebecca Lowe Sandra Flynn


1. Context2. Methods3. What have we found and has it

had any impact?4. What else can we do?

1. Context

Context

• 4,500 suicides per year in England• Suicide is a major contributor to

premature mortality (in men 2nd to heart disease)

Trend in suicide rate for young men (aged 20-34)Death rates from Intentional Self-harm and Injury of Undetermined Intent, England

0

5

10

15

20

25

30

1970 1975 1980 1985 1990 1995 2000 2005

Age standardised death rateper 100,000 population

Males 20-34

Rates are calculated using population estimates based on 2001 census. Rates are calculated using the European Standard Population to take account of differences in age structure.

Years to 1998 and 2000 have been coded using ICD9; 1999 and 2001 onwards are coded using ICD10.

Source: ONS (ICD9 E950-E959, plus E980-E989, excluding E988.8 (inquest adjourned) ; ICD10 X60-X84, Y10-Y34 excl. Y33.9 (verdict pending))

All Persons, All Ages

Fig 4:

Three-year average rate, plotted against middle year of average (1969-2006)

Context

• 4,500 suicides per year in England• Suicide is a major contributor to

premature mortality in England • Latest annual rate: 8.5 per

100 000 population

Suicide in mental illness

• Psychological autopsy studies• Cohort studies

Psychological autopsy studies

11%

35%

22%

12%

6%

14%

Schizophrenia

Mood Disorders

SubstancedisordersPersonalitydisordersAnxietydisorders

Other disorders

(Bertolote & Fleischmann 2002)

Cohort studiesDisorder SMR

Schizophrenia 845 Bipolar disorder 1505 Major depression 2035Dysthymia 1212 Panic disorder 1000Alcohol misuse 586

Suicide in mental illness

2. The National Confidential Inquiry - Methods

The National Confidential Inquiry

• A UK wide study• Based in Manchester since 1996• Collects national data on suicide and

homicide in the general population and more detailed data on those under the care of mental health services at the time of death

• Recommends changes to practice and policy


• All ‘suicide deaths’ from National Statistics

• Include both suicide and undetermined verdicts

MethodsObtain national data

Determine contact with MH services via NHS Trust/board contact

No contact with services<12 months

Contact with services<12 months

Questionnaire sent to consultant


Questionnaire:• 25 pages• 11 sections• Demographic details• Clinical details• Details of management

• Response rate 97%


Why is our response rate so high?

• Been around since 1996• Procedures and mechanisms well honed

(e.g. reminder system)• System of trust contacts

The National Confidential InquiryWhy is our response rate so high?

• Clinicians expect the questionnaires• Clinicians see the value of the data• No blame• No individual analysis of cases or

presentation of identifiable data• Part of arrangements for clinical

governance (England only)

The National Confidential InquiryApprovals

• Ethics• Data protection and security• Section 60 approval (allows processing

of identifiable information without explicit consent)

The National Confidential InquiryDatabases

• Operating for 12 years

• 70,000 individuals who have died by suicide on the general population database.

• Detailed clinical data on over 17,000 individuals on the Inquiry database

3. What have we found and has it had any impact?

The National Confidential InquiryEngland and Wales 2001-2004• General population suicide deaths:

23,477• Rate: 10.2 per 100,000 per year• Inquiry cases: 6,397 (27%)• 1300 deaths per year

Timing of last contact

1153

1802

1430

830 865

0

200

400

600

800

1000

1200

1400

1600

1800

2000

< 24 hours 1-7 days 1-4 weeks 5-13 weeks >13 weeks

Timing of last contact

Freq

uenc

y

0

200

400

600

800

1000

1200

1400

1600

1800

2000

Hanging/st rangulat ion Self -poisoning Carbon monoxidepoisoning

Jumping/mult ipleinjuries

Drowning Other

Cause of death

Freq

uenc

yMale Female

Method of suicide for Inquiry cases

Inquiry cases• High levels of social isolation, self-harm

and substance misuse• 856 (14%) in-patients• 1271 (20%) died within 3 months of

discharge• 1523 (29%) missed last contact with

services• 4984 (86%) immediate risk estimated as

low or absent• 1017 (19%) thought to be preventable

Inquiry casesIn-patients• 27% of deaths occurred after patients

had left the ward without staff permission

• 22% of deaths occurred in patients under non-routine observation

Post discharge deaths• Timing

Post-discharge deaths

68675876687374

88

111113

145138

192

0

25

50

75

100

125

150

175

200

225

1 2 3 4 5 6 7 8 9 10 11 12 13

Weeks between discharge and suicide (Week 1 = First week following discharge)

Freq

uenc

y

What else might services do?

In-patients• Reduce absconding• Improve observation protocols• Make wards safer

Community patients• Manage the transition from ward to

community

What else might services do?Number

(188) %Ligature Type:Belt 82 45%Sheet, towel, etc. 29 16%Shoelaces 18 10%Clothing (tie, scarf, tights, etc.) 19 10%Item brought in specifically (e.g. rope) 4 2%Other specified (e.g. cable cord, curtains) 29 16%

Ligature Point:Hook or Handle 42 23%Door 32 18%Window 23 13%Bed head 18 10%Other rail (e.g. toilet rail, wardrobe rail) 10 6%Pipes 9 5%Shower fixtures (e.g. shower head, tap) 8 4%Bed curtain rail 6 3%Other specified (e.g. light fixture, radiator) 33 18%

What else might services do?

In-patients• Reduce absconding• Improve observation protocols• Make wards safer

Community patients• Manage the transition from ward to

community

Has the Inquiry had an impact?

0

2

4

6

8

10

12

14

1997 1998 1999 2000 2001 2002 2003 2004

Year

Sui

cide

rate

per

100

,000

pop

ulat

ion

general population suicide rate per 100,000 populationInquiry suicide rate per 100,000 population


• In-patients


Rate of in-patient suicide, England 1997-2003

0

1

2

3

1997 1998 1999 2000 2001 2002 2003

Year of death

In-p

atie

nt ra

te p

er 1

00,0

00 b

ed d

ays

In-patient


Rate of in-patient and post-discharge suicide, England 1997-2003

0

1

2

3

1997 1998 1999 2000 2001 2002 2003

Year of death

In-p

atie

nt ra

te p

er 1

00,0

00 b

ed

days

0

2

4

6

8

10

12

14

16

Post

-dis

char

ge ra

te p

er 1

0,00

0 ad

mis

sion

s

In-patientPost discharge


Priority groups• Psychiatric in-patients• Those who die within 3 moths of

discharge from in-patient care• Those under CPA• Those who are non-compliant at the

time of death• Those who missed their last

appointment with services


-60

-40

-20

0

20

40

60

% c

hang

e in

sui

cide

from

199

7 to

200

4

0 1 2 3

Number of priority groups that an individual belongs to

4+


187190

149155

249

289

239258

0

50

100

150

200

250

300

350

1997 1998 1999 2000 2001 2002 2003 2004

Year

Freq

uenc

yNumber of deaths following loss of treatment


Inquiry recommendations: • Removal of ligature points• Assertive community teams• Risk management training• Individual care plans• Easy access in crisis• Services for dual diagnosis patients• Post-discharge follow up• Information sharing• Post-incident review


0

10

20

30

40

50

60

70

80

0 1 2 3 4 5 6 7 8 9

Number of implemented recommendations in 1998

Num

ber o

f Tru

sts

05

1015202530354045505560657075

1998 1999 2000 2001 2002 2003 2004 2005 2006

Year

Num

ber

of T

rust

s

0 KSRs 1 KSR 2 KSRs 3 KSRs 4 KSRs 5 KSRs 6 KSRs 7 KSRs

8 KSRs 9 KSRs


Policy impact of the NCI

• Definitive figures on suicide• Contributed to the National Service

Framework (NSF) Standard 7 (suicide prevention)

• Clinical recommendations in the NHS plan (e.g. assertive outreach teams, and improving access in crisis)

• Safety standards adopted by the NHS Clinical Negligence Scheme for Trusts (CNST)

Policy impact of the NCI

• Specific recommendations on patient safety, such as: the removal of ligature points on in-patient wards, early follow-up of post-discharge psychiatric patients.

• A safety checklist for mental health services incorporated into the National Suicide Prevention Strategy (NSPS) (“12 points to a Safer Service”)

• Data to individual NHS trusts to support clinical governance

4. What else can we do?

Controlled studiesPsychiatric in-patients: a case control study

• 23% died within the first week of admission

• Risk factors included self-harm, life events, symptoms at last contact, more than one psychiatric diagnosis, being off the ward without staff agreement.

• Unemployment was protective

Contact with other servicesGP Contacts

Contacts with the GP in the 12 months prior to death

43373126231715131197531

Freq

uenc

y

20

10

0

Emergency Department Contacts

Contacts with ED in the 12 months prior to death

3017111097654321Fr

eque

ncy

30

20

10

0

Contact with other services“Always a feeling of desperation in (GP area), that whatever you try to get sorted always seems to be stonewalled by whoever you try to get through to. So knowing that he felt suicidal and trying to get him seen… and he knew that it was going to be a waste of time and we knew that it was going to be a waste of time, so together we were feeling a feeling of desperation”

Trends

00.5

11.5

22.5

33.5

44.5

55.5

6

1997

1998

1999

2000

2001

2002

2003

Rat

e/10

0,00

0OverallMalesFemales35%

28%

Suicide rates among young people in the UK

Subgroups

Young people • Suicide rates higher in males, higher in 15-

19 year olds• Low rate of contact

– 14% (overall)– 12% (males)– 20% (females)

Ethnic minority groups• Suicides characterised by violent methods,

schizophrenia, recent non-compliance, previous violence, unemployment

Other outcomes

• Homicide• Sudden Unexplained Deaths on

psychiatric wards• Overdose deaths and medical

management

East Sussex, Brighton and Hove

Avon

Suicide away from home

• 12% all suicide deaths are ‘away from home’

• These individuals: younger, homeless, unemployed, BME, SCZ, inpatient status

• Methods: CO, drowning, jumping

• Location was associated with method of death – e.g. drowning in coastal areas, jumping among non-residents in hot spot areas

Cross-national comparisons

Scotland

• 5,054 general population suicides - 18.7 per 100,000 population per year

• 1,373 patient deaths (Inquiry cases)• 28% of all suicide deaths in Scotland• 229 per year

Rate of in-patient suicide

0.72

1.28

0.970.94

1.38

1.061.2

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1998 1999 2000 2001 2002 2003 2004

Year

Rat

e pe

r 100

,000

be

d da

ys

Rate of post-discharge suicide

11.7

17.618.4

22.3

1717.116.5

0

5

10

15

20

25

1998 1999 2000 2001 2002 2003 2004

Year

Rat

e pe

r 100

,000

be

d da

ys

Comparisons with England and Wales

Scotland E & W

Suicide rate per 100,000 population 18.2 vs. 10.2

(Schizophrenia 0.79 vs. 0.53)

Rate of contact 28% vs. 23%

Self-poisoning 34% vs. 28%

Drowning 10% vs. 6%

Low/no immediate risk 91% vs. 86%

Preventability 11% vs. 19%

In-patients 9% vs. 14%

General population suicide profile for Scotland compared to England and Wales

02

46

81012

1416

182022

2426

2830

11 13 15 17 19 25-30

35-39

45-49

55-59

65-69

75-79

85-89

Age bands

Scotland

E W 2000-2005


Scotland E & W

Suicide rate per 100,000 population 18.2 vs. 10.2

(Schizophrenia 0.79 vs. 0.53)

Rate of contact 28% vs. 23%

Self-poisoning 34% vs. 28%

Drowning 10% vs. 6%

Low/no immediate risk 91% vs. 86%

Preventability 11% vs. 19%

In-patients 9% vs. 14%

17

58

9

39

8

44

3

30

0

10

20

30

40

50

60

70

Alcoholdependence

Alcohol misuse Drug dependence Drug misuse

%

Scotland E & W


Recommendations

Community services

• Better training and services for the management of drug and alcohol misuse – including dedicated services for dual diagnosis

• Specialist community mental health teams providing outreach for patients at risk of losing contact

• Early follow-up following hospital discharge

Recommendations

In-patient units

• More intensive supervision of patients recently admitted to hospital

• Removal of ligature points from in-patient wards

• Prevention of absconding from wards

• Careful assessment of risk during periods of leave leading up to discharge

Recommendations

General

• Positive clinical attitudes to the prevention of risk as part of a more understanding dialogue with the public

• Further study of the higher suicide rates in Scotland


1. Context2. Methods3. What have we found and has it

had any impact?4. What else can we do?

Identifying high risk groups to prevent suicide - The ...nsrf.ie/wp-content/uploads/presentations/Prof Nav Kapur 11-09-14.pdfprevent suicide - The National Confidential Inquiry into

Documents